视觉自我实现对齐：通过威胁相关图像塑造安全导向的人设
Visual Self-Fulfilling Alignment: Shaping Safety-Oriented Personas via Threat-Related Images
AI总结：
本文提出视觉自我实现对齐方法，通过威胁相关图像训练视觉语言模型，塑造安全导向的人设，减少攻击成功率并提升响应质量。
摘要：
多模态大语言模型（MLLMs）面临安全对齐问题，视觉输入可能导致有害输出。为解决此问题，现有方法需显式安全标签或对比数据；然而，威胁相关概念具体且可视觉化，而安全概念如帮助性抽象且缺乏视觉参照。受自我实现机制下涌现对齐的启发，我们提出视觉自我实现对齐（VSFA）。VSFA在围绕威胁相关图像构建的中性VQA任务上微调视觉语言模型（VLMs），无需任何安全标签。通过反复接触威胁相关视觉内容，模型内部化警惕和谨慎的隐含语义，塑造安全导向的人设。在多个VLMs和安全基准上的实验表明，VSFA降低了攻击成功率，提高了响应质量，并缓解了过度拒绝现象，同时保持通用能力。我们的工作将自我实现机制从文本扩展到视觉模态，为VLMs对齐提供了一种无标签的方法。
英文摘要：
Multimodal large language models (MLLMs) face safety misalignment, where visual inputs enable harmful outputs. To address this, existing methods require explicit safety labels or contrastive data; yet, threat-related concepts are concrete and visually depictable, while safety concepts, like helpfulness, are abstract and lack visual referents. Inspired by the Self-Fulfilling mechanism underlying emergent misalignment, we propose Visual Self-Fulfilling Alignment (VSFA). VSFA fine-tunes vision-language models (VLMs) on neutral VQA tasks constructed around threat-related images, without any safety labels. Through repeated exposure to threat-related visual content, models internalize the implicit semantics of vigilance and caution, shaping safety-oriented personas. Experiments across multiple VLMs and safety benchmarks demonstrate that VSFA reduces the attack success rate, improves response quality, and mitigates over-refusal while preserving general capabilities. Our work extends the self-fulfilling mechanism from text to visual modalities, offering a label-free approach to VLMs alignment.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
